The use of artificial intelligence (ai)-based large language models in schools and universities is a great concern. the recently launched large language models have ability generate text or essays just like humans. this model produce outputs which is just like humans as they are trained on very large dataset. plagiarism is student work is also increasing. student use llm to complete their assignments, solving question which affects the learning of students. so, it is important to distinguish between the text generated by human and the text generated by the large language models. that’s why we have invented novelty to distinguish between student written text and ai generated text that is feature engineering. our invented method gives results properly at accuracy of 0.97. it detects the llm generated text and returns accuracy of text. it also gives the length of text, number of words presented in the text and we can give the feedback or also add the votes for accuracy that has been measured by the model. natural language processing (nlp) plays a crucial role in the functionality of llms, allowing them to interpret human language, adapt to individual learning styles, and provide constructive feedback. however, the reliance on llms for educational tasks poses challenges, as the ease with which students can generate content raises questions about the authenticity of their work and the effectiveness of traditional assessment methods. it will help us improve how we recognize text generated by large language models (llms) by creating machine learning algorithms that can identify different features of llm-generated content in student submissions. so, we have built ui for our text detection.